Molecular Cancer BioMed Central Research Open Access G-Catenin promotes prostate cancer cell growth and progression by altering cell cycle and survival gene profiles Yan Zeng1, Agustin Abdallah1, Jian-Ping Lu1, Tao Wang1,3, Yan-Hua Chen1,2, David M Terrian1,2, Kwonseop Kim1,4 and Qun Lu*1,2 Address: 1Department of Anatomy and Cell Biology, Brody School of Medicine, East Carolina University, Greenville, NC 27858, USA, 2Leo Jenkins Cancer Center, Brody School of Medicine, East Carolina University, Greenville, NC 27858, USA, 3Department of Surgery, Beijing Capital Medical University, Beijing, PR China and 4College of Pharmacy, Chonnam National University, Gwangju, Republic of Korea Email: Yan Zeng - [email protected]; Agustin Abdallah - [email protected]; Jian-Ping Lu - [email protected]; Tao Wang - [email protected]; Yan-Hua Chen - [email protected]; David M Terrian - [email protected]; Kwonseop Kim - [email protected]; Qun Lu* - [email protected] * Corresponding author Published: 10 March 2009 Received: 17 October 2008 Accepted: 10 March 2009 Molecular Cancer 2009, 8:19 doi:10.1186/1476-4598-8-19 This article is available from: http://www.molecular-cancer.com/content/8/1/19 © 2009 Zeng et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Background: G-Catenin is a unique member of E-catenin/armadillo domain superfamily proteins and its primary expression is restricted to the brain. However, G-catenin is upregulated in human prostatic adenocarcinomas, although the effects of G-catenin overexpression in prostate cancer are unclear. We hypothesized that G-catenin plays a direct role in prostate cancer progression by altering gene profiles of cell cycle regulation and cell survival. Results: We employed gene transfection and small interfering RNA to demonstrate that increased G-catenin expression promoted, whereas its knockdown suppressed prostate cancer cell viability. G-Catenin promoted prostate cancer cell colony formation in soft agar as well as tumor xenograft growth in nude mice. Deletion of either the amino-terminal or carboxyl-terminal sequences outside the armadillo domains abolished the tumor promoting effects of G-catenin. Quantitative RT2 Profiler™ PCR Arrays demonstrated gene alterations involved in cell cycle and survival regulation. G-Catenin overexpression upregulated cyclin D1 and cdc34, increased phosphorylated histone-H3, and promoted the entry of mitosis. In addition, G-catenin overexpression resulted in increased expression of cell survival genes Bcl-2 and survivin while reducing the cell cycle inhibitor p21Cip1. Conclusion: Taken together, our studies suggest that at least one consequence of an increased expression of G-catenin in human prostate cancer is the alteration of cell cycle and survival gene profiles, thereby promoting tumor progression. Background E-catenin and adenomatous polyposis coli (APC), con- Tumor progression is the result of loss of balance in cellu- tribute to cancer development by disrupting the E-cad- lar functions including cell growth, adhesion, division herin based cell-cell junction, as well as interfering with and apoptosis. Among the many oncogenes and tumor cell proliferation, altering karyotype, and reducing apop- suppressors, cell-cell junction associated proteins, such as tosis [1,2]. Page 1 of 11 (page number not for citation purposes) Molecular Cancer 2009, 8:19 http://www.molecular-cancer.com/content/8/1/19 G-Catenin, or NPRAP (neural plakophilin related arma- a bone metastasis of prostatic adenocarcinoma [15], were dillo protein)/Neurojungin, is an adhesive junction asso- chosen to test the hypothesis that the increase or decrease ciated protein [3,4], which was initially identified as a in G-catenin expression affect prostate cancer cell growth, neural specific protein [5,6]. G-Catenin belongs to the respectively. p120ctn subgroup in the armadillo/E-catenin superfamily [5,7]. While E-catenin and p120ctn are ubiquitously We transfected G-catenin cDNA with EGFP fusion into expressed in the body, G-catenin distribution is principally CWR22Rv-1 cells (Fig 1A). The stable cell lines were estab- restricted to the brain in healthy individuals. However, it lished by G418 selection followed by cell sorting to enrich has become increasingly clear that G-catenin is expressed for GFP emitting cell populations. This method allowed in a variety of cancers of peripheral tissues, including for the establishment of cell culture that contained almost breast, prostate, and esophageal tumors [8,9]. Recently, 100% G-catenin overexpressing cells. While vector trans- we showed that G-catenin is upregulated in over 80% of fected cells showed clear monolayer cell morphology (Fig prostatic adenocarcinomas, and its expression is corre- 1A, a), G-catenin overexpressing cells tended to form clus- lated with increasing Gleason scores [9]. An increased ters (Fig 1A, d). This result was reminiscent of the disrup- expression of G-catenin is accompanied by reduced E-cad- tion of cell monolayer morphology in MDCK kidney herin and p120ctn in primary prostatic adenocarcinomas, epithelial cells overexpressing G-catenin [3]. Vector trans- and the forced overexpression of G-catenin in cultured fected cells, with GFP as the transfection marker uni- prostate cancer cells can induce the redistribution of E- formly distributed in the cells (Fig 1A, b; see insert), cadherin and p120ctn [9]. While it is established that cell- displayed E-cadherin at the cell-cell junction (Fig 1A, c; see cell junction proteins, such as E-cadherin, E-catenin, APC, insert). However, G-catenin overexpression at the cell-cell and p120ctn, are involved in cell adhesion and motility as junction (Fig 1A, e; see insert) showed disrupted E-cad- well as cancer cell growth [1,2,10,11], it is not clear herin distribution (Fig 1A, f; see insert). whether G-catenin overexpression exerts any effects on prostate cancer cells. A number of studies showed that disrupted cell-cell junc- tion can alter cell growth [10,11]. To investigate the effects In this study, we tested the hypothesis that G-catenin plays of G-catenin expression in prostate cancer cells, we exam- a direct role in prostate cancer cell growth by altering gene ined the growth property of CWR22Rv-1 cells stably over- profiles of cell cycle regulation and cell survival. We dem- expressing G-catenin or stably suppressing G-catenin onstrated, for the first time, that G-catenin overexpression expression (Fig 1B). To determine if the endogenous G-cat- promotes anchorage-independent prostate cancer cell enin expression plays an important role in cell growth, growth and tumor xenografts in nude mice. We deter- stable cell lines expressing small hairpin RNAs (shRNAs) mined that the ability of G-catenin overexpression to pro- specific for G-catenin gene were established and confirmed mote prostate tumor xenograft growth is dependent on by Western blots (Fig 1B, a). Compared with vector trans- the amino- (NH2) and carboxyl- (COOH) terminal fected cells, G-catenin overexpression showed a significant sequences flanking the armadillo repeat domains. In addi- increase in cell numbers (Fig 1B, b, compare G-catenin tion, quantitative RT2 Profiler™ PCR arrays revealed a wide with vector). Compared with vector transfected cells, two range of gene alterations involved in cell cycle and sur- independent shRNAs against different G-catenin sequences vival regulation. These findings support the notion that at reduced viable cell numbers (Fig 1B, b, compare vector 1 least one consequence of an increased G-catenin expres- and 2 with shRNA 1 and 2). sion in prostate cancer development is the alteration of cell cycle and survival gene profiles, thereby promoting To determine if the growth promoting effects of G-catenin tumor progression. expression also applies to other cancer cell types, we examined PC-3 prostate cancer cells and NCI-H1299 cells Results derived from a human lung carcinoma [16]. Similarly, -Catenin overexpression promotes prostate cancer cell viable cell numbers were significantly increased in PC-3 growth in culture and NCI-H1299 cells stably overexpressing G-catenin (Fig Our previous studies showed that G-catenin expression is 1B, c and 1B, d). These observations indicated that G-cat- very weak in normal prostatic glandular epithelial cells enin expression is important for cancer cell growth in cul- but is remarkably increased in prostatic adenocarcinoma ture. [9]. Screening for prostate cancer cell lines overexpressing G-catenin, we found that G-catenin expression was moder- -Catenin promotes the anchorage-independent growth of ately increased in CWR22Rv-1 and PC-3 cells when com- CWR22Rv-1 cells in soft agar and tumor xenografts in pared to non-cancer prostate epithelial cells PZ-HPV-7 but nude mice remained very low in LNCaP and DU145 cells [12]. There- To investigate whether G-catenin overexpression promotes fore, CWR22Rv-1, derived from a recurrent human pros- the anchorage-independent growth of prostate cancer tate cancer xenograft [13,14], and PC-3 cells, derived from cells, we performed soft agar assays. Following the plating Page 2 of 11 (page number not for citation purposes) Molecular Cancer 2009, 8:19 http://www.molecular-cancer.com/content/8/1/19 FigureG-Catenin 1 expression is important for viable prostate cancer cell growth G-Catenin expression is important for viable prostate cancer cell growth. A. Establishment of stable CWR22Rv-1 cells overexpressing G-catenin and its effects on epithelial cell morphology. CWR22Rv-1 cells, showing epithelial morphology (a) when transfected with pEGFP as vector control (b), display expression of E-cadherin at the cell-cell junction (c). G-Catenin overexpressing CWR22Rv-1 cells (e) interfere with the epithelial monolayer (d) and disrupt E-cadherin expression at the cell- cell junction (f). Bar, 30 Pm. Inserts: selective higher magnification images for (b), (c), (e) and (f), respectively.